Incorporating Test Design into Factorial Validity: A Bifactor Reanalysis of the BAFACALO Folding Test
REMUNOM · 06.10.2026 · 13 (21) · 1-17
DOI: 10.66104/00fx1964 ↗Kurz gefasst
Fachartikel in der Zeitschrift REMUNOM, online erschienen am 06.10.2026. Verfasst von Cristiano Mauro Assis Gomes und 1 weiteren Autor:innen. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 57 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
Evidence-based psychological assessment requires well-calibrated evidence about the dimensional structure of tests, since every score interpretation presupposes that structure. The open science movement has established test democratization as an important agenda, and by broadening the use of tests, democratization makes well-calibrated structural evidence even more important. The Folding Test, part of the BAFACALO battery, has been freely accessible since 2021, and its factorial structure was investigated through an exploratory strategy with the estimation procedures available at the time, yielding a general factor (visualization) and three specific factors. Because the 30 items of the test are organized into six figures, these specific factors may arise from the format of the test itself. The present study reexamines the dimensional structure of the Folding Test through a confirmatory factor analysis with 292 Brazilian public high school students. Three confirmatory factor models were compared using WLSMV estimation, a procedure specific to dichotomous items: a unidimensional model, a bifactor model derived from a previous exploratory study of the test, and a bifactor model with one specific factor per figure (figure model). All models showed acceptable fit, and both bifactor models fitted slightly better than the unidimensional model. After adjustment, the two bifactor models converged to practically the same structure, with specific factors centered on items of figures 1 and 3, and could not be distinguished by fit. The figure model was retained for its theoretical pertinence (scaled CFI = .997; scaled RMSEA = .032). The visualization factor showed high reliability (McDonald's ω = .979), whereas the specific factors showed low reliability (ω ≤ .349). The Folding Test is essentially unidimensional, and its specific factors are better explained by the test format than by distinct abilities. The study complements the democratization of the Folding Test with better-calibrated evidence about its dimensional structure.
Zitieren
APA
Gomes, C. M. A., & Araujo, J. D. (2026). Incorporating Test Design into Factorial Validity: A Bifactor Reanalysis of the BAFACALO Folding Test. REMUNOM, 13(21), 1-17. https://doi.org/10.66104/00fx1964
Vancouver
Gomes CMA, Araujo JD. Incorporating Test Design into Factorial Validity: A Bifactor Reanalysis of the BAFACALO Folding Test. REMUNOM. 2026;13(21):1-17. doi:10.66104/00fx1964
BibTeX
@article{Gomes2026,
author = {Gomes, Cristiano Mauro Assis and Araujo, Jhonys de},
title = {Incorporating Test Design into Factorial Validity: A Bifactor Reanalysis of the BAFACALO Folding Test},
journal = {REMUNOM},
year = {2026},
volume = {13},
number = {21},
pages = {1--17},
doi = {10.66104/00fx1964}
}
EUVIMED